"John" <[EMAIL PROTECTED]> wrote > I've now written my first set of Classes to do some fairly specific > processing for work I do. I have a few questions. > > First, in looking through what I've done, I basically just > incorporated all > my previous scripts into classes... they are still highly specific > to my > application, though I did try to make them somewhat 'reusable' or > general.
That may be OK but without knowing more about what these scripts do then its hard to say. Its possiblye that the scripts are all of the same general nature and deal with comon data elements or perform similar functions but with different algorithms or data parameters. In those cases there are probably ways to use abstract classes to create a uniform class framework that can then be specialised for each task. But to know if that makes sense requires a lot more information about the nature of the problem Also, it depends a bit on how you implemented the classes. If you literally just put class foo: at the top of the module and treated all variables as attributes and converted all functions to methods then probably you haven't gained much advantage. > I guess, I'm writing because I'm wondering now what people think > about > writing classes versus just using scripts for things that are so > specific. The more specific the task the less useful clases will be. As the name suggests classes are used for solving problems that can be classified, wjhere the generic aspects can be captured in one place (a superclass) and the specifics put someplace else (a subclass). > One thing I am struggling with is how to assign *args and **kwargs > if they > are passed, and how to ignore them if they are not... right now I do > this: > > def myfunc(self, *args,**kwargs): > a=self.a > b=self.b > kwa=self.kwa > kwb=self.kwb > try: > a=args[0]; b=args[1] > kwa=kwargs['a'] > kwb=kwargs['b'] > except: pass > > > Where self.X is defined in the __init__ of the class. Is that > correct? It should work although in my (limited) experience of using *args/**kwargs it is more common for them to be used in this way within the init() itself rather than in another method. When you create an instance you would pass all the attriubute data it needs to do its job. Once instantiated the methods of the object should rarely need to set lots of attributes to new values, maybe one or two but not many. Typically that suggests that some other object is messing with data that the instance should be managing itself! But again without context its impossible to be definitive, there are exceptions to every rule - especially in OOP! But the pattern would generally raise alarm bells for me until I understood why it was that way. HTH, -- Alan Gauld Author of the Learn to Program web site http://www.freenetpages.co.uk/hp/alan.gauld _______________________________________________ Tutor maillist - Tutor@python.org http://mail.python.org/mailman/listinfo/tutor
